Kimi K2 0905 vs Kimi K3.1 Pro: Specs & Benchmark Comparison

Kimi K2 0905 and Kimi K3.1 Pro both come from Moonshot AI, Kimi K2 0905 was released in September 2025, and Kimi K3.1 Pro followed 13 months later in October 2026. Kimi K2 0905 has a published size of about 1.0 trillion parameters; Moonshot AI has not disclosed the parameter count of Kimi K3.1 Pro.

We have 6 benchmark results for Kimi K2 0905 and 31 benchmark results for Kimi K3.1 Pro, but they overlap on a single test: GPQA, where Kimi K3.1 Pro scores 94.2% against 75.8%.

Kimi K2 0905 is the cheaper API at $0.60 per million input tokens and $2.50 per million output tokens, roughly 3 times cheaper than Kimi K3.1 Pro at $2 and $10. Kimi K3.1 Pro takes the larger context window at 1M tokens, compared with 262K for Kimi K2 0905. Kimi K2 0905 accepts text as input, while Kimi K3.1 Pro accepts text and images. On tooling, only Kimi K2 0905 supports function calling and only Kimi K2 0905 offers structured output.

CharacteristicKimi K2 0905Kimi K3.1 Pro
CompanyMoonshot AIMoonshot AI
Release DateSeptember 5, 2025October 23, 2026
Parameters1.0T—
MultimodalNoYes
Context (input)262K1.0M
Context (output)262K1.0M
Input Price / 1M$0.60$2.00
Output Price / 1M$2.50$10.00
Average Score84.0%65.2%
Benchmarks
GPQA75.8%94.2%

Visual Benchmark Comparison

Kimi K2 0905
Kimi K3.1 Pro
GPQA0.8 vs 0.9
0.8
0.9

Verdict

Kimi K3.1 Pro leads in 2 out of 4 comparison categories.

Overall Performance

Both models show comparable average scores: Kimi K2 0905 — 0.8, Kimi K3.1 Pro — 0.7.

API Cost

Kimi K2 0905 is 3.9x cheaper: input $0.60/1M vs $2.00/1M tokens.

Context Window

Kimi K3.1 Pro supports a larger context: 1M vs 262K tokens.

Recency

Kimi K3.1 Pro is newer: released 10/23/2026 vs 9/5/2025.

More About These Models

Related Comparisons

Frequently Asked Questions

Which is better for coding — Kimi K2 0905 or Kimi K3.1 Pro?
Direct comparison on the SWE-Bench benchmark is not available. We recommend reviewing other metrics on the comparison page.
Which model is cheaper — Kimi K2 0905 or Kimi K3.1 Pro?
Kimi K2 0905 is cheaper for input: $0.60 per 1M tokens vs $2.00.
Which has a larger context window — Kimi K2 0905 or Kimi K3.1 Pro?
Kimi K3.1 Pro supports a larger context: 1,048,576 tokens vs 262,144.

The Kimi K2 0905 and Kimi K3.1 Pro comparison is updated for 2026. Data includes benchmark results, API pricing, context window size and other specifications. For more detailed information, visit the Kimi K2 0905 or Kimi K3.1 Pro page. See also the complete list of AI model comparisons.